We see some glitch after each conversion. Some conversions seem to generate large glitches (100mV or more). We use bias compensation - and a capacitor of 10pF or more does not do any difference.
I use a filter after the output cutting at 20kHz. Still I see some residuals from the glitches. What can be done? Cut off at lower frequencies?
Are you using an evaluation board(EVAL-AD5791SDZ)? If not, would you be able to share a schematic of the circuit that you are using? Is this happening on a single code, low codes, mid codes, high codes, all codes? Is it repeatable? Would you have some scope shots available?
Hereby a snapshot of the schematic.
Actually I don’t know the codes at which glitches are worst at. My assumption is it happens at certain codes.
The blue trace shows the IC12B-pin7. The red trace shows IC12A-pin1.
Kind regards Martin
We use AD688 as reference-ic for this DAC. Is that too bad?
Apologies for the delay. You mentioned the connection of the 10pF bias capacitor. Where exactly is this capacitor connected?
Sorry for my late responce. We are working on the AD5791 again in another project. This time we need to avoid the glitch noise.
I attach the scematic. I have tried implementing af de-glich filter but it looks like the ADG333 generates a spike on the pin 18 when its operated. There must be som charge that makes the signal jump.